Encephalomyocarditis (EMC) virus induced diabetes mellitus in mice has similarities to a subset of juvenile onset diabetes in humans. Two otherwise similar variants derived from the M- strain of EMC virus are available for study that differ in their interferon inducing particle (ifp) phenotype and in their ability to cause diabetes in mice. The EMC-B variant is ifp+, produces high levels of circulating interferon (IFN) in infected mice, causes limited infection of pancreatic islets, and does not cause diabetes. When circulating IFN is neutralized by anti-IFN globulins, diabetes results from infection with EMC-B indicating that the interferon system, possibly the ifp phenotype, is a determinant of the diabetic outcome. The EMC-D variant is ifp-, produces lower levels of circulating IFN, destroys pancreatic islets and causes diabetes in mice. Key genetic differences must be responsible for the different biological properties of these closely related viruses. In order to determine the genetic basis for these properties we have cloned cDNA corresponding to the complete open reading frame of both variants. Their restriction endonuclease maps are identical for 12 different enzymes. Also, the first 103 nucleotides of the untranslated RNA at the 5' end of the genomes are identical. We propose to discover the genetic basis for the ifp phenotype and diabetogenicity of EMC virus by first comparing the complete nucleotide sequences of the B and D variants. The functional significance of differences in nucleotide sequence that are found to occur in the open reading frame will be ascertained by reference to the in vitro translation products of RNA transcripts and to the known functions of EMC viral proteins. Complete cDNA of the B and D variants will be synthesized for transfection of mammalian cells. With reference to differences in nucleotide sequence within or outside the open reading frame, chimeric molecules of B and D will be synthesized. The biological properties of the recombinant viruses will be determined in order to independently determine the genetic locus responsible for the ifp phenotype and diabetogenicity. This knowledge will further our understanding of virus virulence and the mechanism of interferon induction by viruses. Using data provided by this project, it may be possible to predict the virulence of human viruses from a knowledge of genetic structure.